Application of HAH Model in Cyclic Simple Shear at Large Accumulated Strain

Abstract

In sheet metal forming, several strain path changes can occur during the process and lead to a large accumulated strain. As the deformation history may involve cyclic loading at large strain, the commonly used uniaxial tensile test is insufficient to capture the hardening behavior correctly. In this study, simple shear tests involving large accumulated strain and several cycles have been performed on a dual-phase steel sheet, namely, DP780. Since this material is an advanced high strength steel (AHSS), the tests were performed on a simple shear test device specifically built for this type of materials. Three simple shear tests with one, two and three cycles were conducted until the same 100% accumulated shear strain was obtained. Then, a distortional plasticity model, so-called HAH, was employed to predict the stress-strain curves obtained during these three tests. The HAH model was considered with the Yld2000-2d yield function to characterize plastic anisotropy and the Hockett-Sherby law as the monotonic hardening description in this study. In a first examination, the optimized HAH coefficients were obtained from the tension-compression test calibrated with a single and multiple cycle tests. Generally, a good agreement was obtained between predicted and experimental stress-strain curve for low strains or a most for two cycles. However, since discrepancies were observed are larger strains, other hardening laws, namely, a modified Voce law with an added linear term, and a Swift-modified Voce law were tested to improve the predictions. However, although the results were slightly improved with the additional hardening laws, it was shown that the permanent softening was not captured accurately at large accumulated strain. This work shows that the permanent softening is not strain dependent only but varies as the number of cycles involved increases, a behavior that cannot be predicted accurately because this feature is not available in the HAH model.